Aim 1: We found that proteolytic cleavage dictates different cellular localization and functions of p12. The removal of a non-canonical endoplasmic reticulum (ER) retention/retrieval signal within the amino terminus of p12 is necessary for trafficking to the Golgi apparatus and generation of a completely cleaved 8kD protein. The 8kD protein in turn traffics to the cell surface, is recruited to the immunological synapse following T-cell receptor (TCR) ligation and down-regulates TCR proximal signaling. The uncleaved 12kD form of p12 resides in the ER and interacts with the beta and gamma chains of the interleukin-2 receptor (IL-2R), the heavy chain of the major histocompatibility complex (MHC) class I, as well as calreticulin and calnexin. Genetic analysis of ORF-I from ex vivo samples of HTLV-I-infected patients reveals predominant amino acid substitutions within ORF-I that inhibit proteolytic cleavage, suggesting that ER associated functions of p12I may contribute to the survival and proliferation of the infected T-cells in the host Our previous work suggested that the effects of the ORF-I proteins on the IL-2 R, STAT-5, and MHC-I occur in a pre-Golgi compartment and is likely mediated by the 12kD form. However, because those experiments were performed with an ORF-I cDNA that also encoded the 8kD form, we need to evaluate whether the 8kD form affects surface expression of the IL-2 R MHC-I, ICAM1 and ICAM34 as well. We have generated lentivirus vectors that express both ORF-I isoforms individually and plan to characterize the effect of these proteins on the expression and recycling of these receptors. We will test whether this effect is due to the decrease in the phosphorylation of N-WASP, a protein which regulates actin cytoskeleton remodeling, resulting in disruption of cell-spreading and formation of actin-rich circumferential rings following surface-supported TCR stimulation. We plan to determine whether the 8kD protein physically interacts with the TCR and affects the distribution of other T-cell signaling molecules. Our NCI collaborators, F. Ruscetti and K. Jones, have shown that HTLV-I virions can productively infect (cis-infection)monocyte-derived dendritic cells and that DCs exposed to HTLV-I transfer virus to T-cells (trans-fection). Since DCs from infected individuals contain HTLV-I proviral sequences, we wanted to examine the effect of mutations in p12, p13, andp30 on the ability of HTLV-I to infect DCs. Using the in vitro infectivity assay of DCs, we observed that some but not all cell-free virus containing mutations in the non-structural genes efficiently entered monocyte-derived dendritic cells. In three independent experiments, we observed that ablation of p12 and p30, but not p13 or HBZ, dramatically decreases the ability of HTLV-I to productively infect dendritic cells. We are currently examining whether viruses carrying these mutations are also reduced in their ability to infect plasmacytoid DCs and whether any of the mutations affect the DC-mediated infection of T-cells. In future experiments, we plan to characterize the block to infectivity of DCs by the viruses carrying mutations in p30 and p12. Based on what is known about infection of other cell types, these two proteins may act at different stages of the viral life cycle during infection of dendritic cells. In macrophages, p30 interferes with TLR4 signaling, stimulates production of IL-10 and decreases the release of TNF-,IL-8 and MCP-1. We will examine whether treatment of DCs with drugs that downregulate TLR4, and/or block TLR4-mediated signaling, restores the infectivity of p30 mutant viruses on these cells in vitro. p12 binds to the vacuolar ATPase which regulates the pH in the endosomes and indeed p12 is found in early and recycling endosomes. p12 may affect the pH in endosomes and thus decrease viral stability and release. This year we reviewed the main functions of the HTLV-1 Orf -I products. The 12 kDa product from orf-I (p12) is proteolytically cleaved within the endoplasmic reticulum (ER) to generate the 8 kDa protein (p8). At the steady state, both proteins are expressed at similar levels in transfected T-cells. The p12 protein remains in the ER and cis-Golgi, whereas the p8 protein traffics to the cell surface and is recruited to the immunological synapse. The p12 and the p8 proteins have seemingly opposite effects on T-cells;the ER resident p12, modulates T-cell activation and proliferation, whereas p8 induces T-cell anergy. The p8 protein also increases the formation of cellular conduits, is transferred to neighboring T-cells, and increases virus transmission. The requirement for HTLV-1 infectivity of orf-I is demonstrated by the loss of virus infectivity in macaques exposed to an engineered virus, whereby expression of orf-I was ablated. Altogether the current knowledge demonstrates that the concerted activity of p8 and p12 is essential for the persistence of virus infected cells in the host Aim 2: Non-structural protein in HTLV-I persistence in the rabbit model, although the rabbit (Leporidae) model of HTLV-I infection has proved very useful to investigate certain aspects of HTLV-I transmission, latency and persistence, it is severely limited by the few reagents available to assess and manipulate the immune response and assess the role of nonstructural HTLV-I proteins in the contest of altered host immune response. Given the cost of studies in non-human primates, we first used rabbits to test the in vivo stability of point mutations introduced into the HTLV-I genome. Because we observed lack of reversion of the point mutations introduced in the HTLV-I mutants in rabbits, we have proceeded to infect Rhesus Macaques (RMs). Rhesus Macaques have been chosen as a model because we have extensive experience in measuring immune responses and manipulating different subsets of immune cells in this animal species. In contrast, all animals exposed to the HBZ knock out virus seroconverted to HTLV-I antigens. Analysis of the virus load levels over time, the extent of viral persistence, and verification of the genotype of the viruses recovered from these animals is ongoing. Macaque model of HTLV-I infection will help address how important the T-cell and B-cell immune response is in the control of this viral infection. The evidence that HTLV-I is controlled by T-cells, though likely remains indirect. In the case that T-cells contribute to the control of virus levels, we will test the effect of CD8 T-cell depletion on provirus load in HTLV-I infected macaques. We observed reversion of the HTLV-1 mutants to the HTLV-1 wild type genotype in none of the inoculated rabbits. In contrast, in macaques, the absence of HBZ was associated with reversion of the mutant virus to the wild type genotype in three of the four animals within weeks from infection and these animals sera recognized HTLV-1 antigens. Similarly, reversion to the wild type was observed in two of the four macaque inoculated with the p30 mutant but only one of these animals fully sero-converted. The four macaques exposed to the p12 knock remained sero-negative and only two animals were positive at a single time point for viral DNA in tissues. Interestingly, we found that the p12 and the p30 mutants were also severely impaired in their ability to replicate in human dendritic cells. These data suggest that infection of dendritic cells may be required for the establishment and maintenance of HTLV-1 infection in primate species. The p30 protein requirement in dendritic cells likely relates to the observation that p30 affects the release of cytokines important for maturation of human macrophage. Further work is necessary to address both of these hypotheses.